Barriers

This section discusses the steps needed to generate the optional fences for the simulations. Preparing a suitable working environment in CloudCompare software is first discussed, showing how to simplify the terrain, and how to place visual cues on the ground to estimate posts spacing and elevation. Then, the steps to generate the base of the fence, the inclination according to the orientation of the terrain and the prerequisites for the simulations are detailed.

Work environment

The design of the barrier measure must be made taking into account several criteria, such as the impact energy, the height of the bounces, the distribution of the structure which must be relatively homogeneous, the accessibility for maintenance... To facilitate this task in 3D, it is advantageous to simplify the terrain for a light computer handeling of the model, detailed in the section “Terrain proxy”. Then, placing visual cues makes it easier to evaluate the site and find the optimal locations. Three types of visual cues are discussed here, in their respective sections: "Topographic contours", "Dot marking" and "Grid marking". Finally, the last section to prepare the working environment shows how to add results of preliminary simulations.

Terrain proxy

This step is only necessary if the terrain model used for the simulations is excessively dense and heavy for the computer to manage. It simply consists in generating a lighter terrain, with a greater spacing of the points. This is done according to the same steps as those presented in the Terrain section. The new terrain generated is then meshed and a white color is assigned to the mesh thus created.

Topographic contours

This step consists in generating visual cues to help orient the fence by following the topographic contours as much as possible. These contour lines can be generated by segmenting the terrain using repeated horizontal slices along the Z dimension using the Cross Section tool. These slices can then be transformed into topographic contour lines on the terrain by calculating the distance between the slices and the terrain, and saturating in the SF display params the colors displayed for the Scalar field of the distances. These colors can then be "saved" by converting them to [R, G, B] values.

Dot marking

This step consists of generating visual cues to facilitate the choice of the spacing of the posts of the fence using points spaced relatively evenly. The method is similar to that of topographic lines to transfer colors to the terrain, by calculating the distances in between the terrain and the points of the dots. The Subsampling tool is used here to first generate the points with the desired spacing.

Grid marking

This step consists in generating visual cues to facilitate the choice of the spacing of the posts of the fence using a grid spaced evenly. The method is similar to the contour lines, but slices in the X and Y dimensions to generate the desired grid.

Addition of preliminary trajectories

Finally, for the preparation of the working environment for the fence, the addition of preliminary trajectories can prove to be very useful when choosing the position of the fence. This short step shows how to add the trajectories to the CloudCompare project. This is done simply by a drag and drop of the trajectories into the project, and skipping the first line of the ASCII file. The same global shift transformation for the transition from georeferenced coordinates to local coordinates must be applied.

The color of the trajectories corresponds to the translational velocities of the simulated rocks, the scale can be displayed via the properties of the layer. These velocities [m/s] can be easily converted into energy [J] if the mass of the rocks is known. This can be done in CloudCompare in two steps with the Scalar fields arithmetics tool (the equivalent of Raster calculator , but for Scalar fields). The first step is to rise the velocity to the power of 2. And the second is to multiply the previous result by half of the mass [kg].

The height of the bounces can be estimated by calculating the distance between the paths of the terrain with the Cloud-to-Cloud Distance tool. The heights obtained do not correspond to the vertical heights of the terrain, however, but rather to the shortest distances between the center of mass of the simulated particles along their path to the terrain. The colors displayed can be adjusted to bring out the bounces exceeding the height of the desired fences, in order to optimize the choice of their location for example.

Tracing base

Now that the work environment is configured, the base of the fence can be traced to the optimal location while taking into account the various prerequisites of the desired fences (most straight deployment possible, easy access, height of the bounces and energy of impacts, etc.). This can be done in CloudCompare using the Trace Polyline tool and clicking on the terrain to indicate the location of the base of each post, across the fence (leaving the option Oversample to 1).

It may be useful to export the coordinates of the points chosen for the base once the effectiveness of the design has been confirmed by simulations and expert judgment. To do this, you must select the vertices of the drawn polygonal line and save them in the desired format (e.g. in shapefile format for later use in a GIS or GPS, or DXF if part of the project requires CAD design). The line of the base can also be exported as a shapefile or DXF if desired, as well as the other elements of the fence generated in subsequent steps.

Terrain orientation

The fences should generally be titled according to the local slope of the terrain. This step therefore consists in evaluating the terrain orientation in order to be able, in the next step, to apply the correct inclination to the fence. The Compass plugin and its Plane tool can be used to assess the local orientation of the terrain. It is necessary to choose a radius which includes a good part of the ground where to install the fence in order to have an average measurement of the ground.

The tool can also be useful for taking structural measurements on cliffs, in order to assess the main discontinuity sets. These can then be used to quickly estimate the types of ruptures that can affect different areas of the terrain with the basic kinematic tests built into stnParabel (see the “Sources” section for more details).

Post angle

Once the terrain orientation has been determined, the angle at which the fence should be inclined can be evaluated according to its specifications (e.g. 19 ° from the vertical for a terrain with a slope of 34 ° for the product RXE-3000 from Geobrugg according to edition 240-N-FO / 06 of their manual). An inclined plane is generated in CloudCompare to be used to generate the net in the next step.

To generate and orient the plan correctly, the orientation plane generated in the previous step can first be duplicated with the Clone tool. The new plane can then be flattened with the Apply Transformation tool. In this tool the dip and dip direction can be entered, the rotation around the center can be applied, and the inverse transform option applied. This brings the plane back to the horizontal. Then a second transformation, not inverted this time, can be applied with a dip corresponding to the desired post angle in the same direction as that of the ground. The new plane should now be properly tilted to the desired orientation of the net. The normal of this plane can be visualized by activating the option Show normal vector, in its properties.

Generating the fence

Once the terrain orientation and the post angle have been evaluated, the fence can be generated. To do this, the base must be duplicated to be used as an upper part. This is then manually moved with the Interactive Transformation Tool by fixing the translation to the Z axis and using the orientation of the previously generated plane. Here, it is necessary to manually move the future top of the fence so that it is positioned beyond the desired height of the net (eg ~10-15 m from the ground, which will subsequently be reduced precisely to the desired height).

The mesh of the fence can then be interpolated between the two polylines representing the base and the top of the fence. To do so, you must select the two lines and use the Mesh/Surface between two polylines tool. The new mesh thus created can be converted into points with the Mesh/Sample points tool using the same point density as that used when creating the terrain. The same goes for the point spacing during downsampling to be applied to the generated points.

Then, to limit the height of the net, the distance of the points generated from the polyline of the base can be evaluated with the Distance to Primitive tool (CloudCompare v.2.11 or +) or from the terrain with the Cloud-to-Mesh Distance tool . The points can finally be segmented to the desired height with the Scalar fileds/Filter by Value tool.

Note: stnParabel v.June 2021 does not consider the deformation of the fence or its ability to resist to the impact. Also, rocks are stopped as soon as they touch the fence, whether their center of mass is inside or beyond the limits of the net, and regardless of the energy at impact. It may therefore be necessary to slightly reduce the height of the fence by half the diameter (d1) of the rocks for the simulations. It may be useful to check the impact energies of the rocks. To do this, the impact points (pos_reach_fence.txt) can be imported in the same way as the trajectories and their translational velocities converted into energy (see the “Addition of preliminary trajectories” section).

Last requirements

Finally, we must ensure that we only have the position of the points for the fence: [X, Y, Z].

Any [R, G, B] colors, surface orientation [Nx, Ny, Nz] and scalar fields must be removed before exporting the fence points in ASCII format.

The ASCII file must be named "fence.txt" or "fences.txt" in order to be correctly taken into account at the same time as the terrain is being loaded for the simulations.

Rendering / other

Here is a small additional section which is not necessary for preparing the files for the simulations, but which can be useful for the presentation of the results and the communication in general. It focuses on the visual rendering of the working environment in CloudCompare related to the fence.